Fan blade inspection method and apparatus based on unmanned-aerial-vehicle aerial photography, and electronic device

Abstract

A fan blade inspection method and apparatus based on unmanned-aerial-vehicle aerial photography, and an electronic device and a computer-readable storage medium, which relate to the technical field of power devices. The inspection method comprises: determining first initial three-dimensional position coordinates and the posture of a fan; on the basis of the first initial three-dimensional position coordinates, the posture and a preset distance, determining second initial three-dimensional coordinates of a first expected position for a front-facing photographic position of the fan; inputting the second initial three-dimensional coordinates of the first expected position into a flight control system of an unmanned aerial vehicle, and on the basis of a GPS system, controlling the unmanned aerial vehicle to fly to the position corresponding to the second initial three-dimensional coordinates; and on the basis of the orientation of the fan, controlling the unmanned aerial vehicle to track, at the position corresponding to actual three-dimensional coordinates and by means of a binocular camera, at least one first key point and at least one second key point of a blade of the fan, and controlling the posture of the camera on the basis of the orientation of the fan and a rotation matrix, so as to capture an image of the fan. The posture of a photographic camera is adjusted by means of binocular vision technology, and the unified orientation can be finally remained to capture an image, thereby improving the quality of the image, and improving the accuracy of blade surface damage inspection.

Description

Wind turbine blade detection method, device and electronic equipment based on drone aerial photography

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This disclosure claims priority to Chinese patent application number 2023112579620 filed with the State Intellectual Property Office of China on September 26, 2023, entitled “Wind blade detection method, device and electronic equipment based on drone aerial photography,” the entire contents of which are incorporated by reference into this disclosure. Technical Field

[0003] The present disclosure relates to the technical field of power equipment, and in particular to a method, device and electronic equipment for detecting wind turbine blades based on drone aerial photography. Background Art

[0004] A wind turbine is an electrical device that converts wind energy into mechanical work, which in turn drives the rotor to ultimately output alternating current. A wind turbine typically consists of blades, a generator, a yoke, a tower, a speed limiter, and an energy storage device. Over the long term, the blades can develop various surface damage, such as damage to the protective film, paint loss, ice formation, cracks, and oil contamination.

[0005] Currently, wind turbine blade surface damage inspections typically use drones equipped with cameras. Since these inspections are image-based, the quality of the images captured directly impacts the accuracy of the inspection results. However, the actual and expected poses of current drones differ significantly when capturing images, significantly impacting image quality and resulting in low accuracy in blade surface damage detection.

[0006] Summary of the Invention

[0007] The present disclosure provides a wind turbine blade detection method, device, and electronic equipment based on drone aerial photography, so as to at least alleviate the technical problem of low accuracy in blade surface damage detection.

[0008] Some exemplary embodiments of the present disclosure provide a wind turbine blade detection method based on drone aerial photography, which may include:

[0009] Determine a first initial three-dimensional position coordinate and attitude of the wind turbine; wherein the first initial three-dimensional position coordinate may be a coordinate in a GPS system, and the attitude may include the orientation of the wind turbine;

[0010] Determining, based on the first initial three-dimensional position coordinates, the posture, and the preset distance, a second initial three-dimensional coordinate of a first expected position for a front shooting position of the wind turbine; wherein the first expected position is located on a first normal vector corresponding to a first rotation center of a blade rotation plane of the wind turbine, and the distance between the first expected position and the first rotation center is the preset distance;

[0011] inputting the second initial three-dimensional coordinates of the first expected position into a flight control system of the drone, and controlling the drone to fly to the second initial three-dimensional coordinates based on the GPS system; wherein the actual position of the drone is the actual three-dimensional coordinate, and the offset between the first expected position and the actual three-dimensional coordinate is a target offset vector;

[0012] Controlling the drone at the actual three-dimensional coordinates based on the orientation of the wind turbine, tracking at least one first key point of a blade of the wind turbine in a first posture using a binocular camera, and determining a three-dimensional motion trajectory of the first key point in a camera coordinate system;

[0013] Determine a second normal vector of a second rotation center corresponding to the three-dimensional motion trajectory of the first key point in the camera coordinate system of the binocular camera, convert the first normal vector to the camera coordinate system to obtain a third normal vector, and determine a first rotation matrix based on the second normal vector and the third normal vector;

[0014] Adjusting the orientation of the binocular camera based on the first rotation matrix so that the actual orientation of the binocular camera is parallel to the normal corresponding to the first rotation center of the actual wind blade rotation plane, obtaining a second posture, and re-tracking at least one second key point of the wind blade using the binocular camera in the second posture to determine the three-dimensional motion trajectory of the second key point in the camera coordinate system and the re-determined second rotation matrix;

[0015] determining whether a difference between the re-determined second rotation matrix and the first rotation matrix satisfies a preset requirement; if so, controlling a camera posture based on the orientation of the wind turbine and the second rotation matrix to capture a front image of the wind turbine; and if not, continuing the iteration until a difference between two adjacent rotation matrices satisfies the preset requirement or the number of iterations reaches a preset number;

[0016] Based on the first initial three-dimensional position coordinates, the posture and the preset distance, the third initial three-dimensional coordinates of the second expected position for shooting the back of the wind turbine are determined, the third initial three-dimensional coordinates of the second expected position are input into the flight control system of the UAV, the UAV is controlled to fly to the optimized position corresponding to the third initial three-dimensional coordinates based on the GPS system, and the camera is controlled to capture the back image of the wind turbine based on the orientation of the wind turbine and the second rotation matrix at the optimized position corresponding to the third initial three-dimensional coordinates.

[0017] In one possible implementation, the drone may be provided with a laser device; and the wind turbine blade detection method may further include:

[0018] When the wind turbine is in a stationary state, the drone is controlled to fly around the blades on the wind turbine based on all relative angles, and during the flight around the blades, the laser device is controlled to always perform multiple laser ranging on the blades by scanning, to obtain multiple laser ranging results, and a current model for the blades is established based on the multiple laser ranging results;

[0019] Comparing the current model with a preset model to obtain a model comparison result; wherein the preset model is a model of the blades of the wind turbine in a normal state before operation;

[0020] The degree of surface damage of the fan blades is detected according to the model comparison result.

[0021] In one possible implementation, the fan blade detection method may further include:

[0022] When the wind turbine is in operation, the drone is controlled to maintain a flight speed relative to the blade that is greater than a rotational speed of the blade, so as to obtain the laser ranging results for multiple positions on the blade.

[0023] In one possible implementation, the number of the drones may be at least two; and the wind turbine blade detection method may further include:

[0024] When the wind turbine is in operation, controlling the first drone to fly at a speed equal to the rotational speed of the blades to keep the first drone and the blades relatively stationary;

[0025] The second UAV is controlled to maintain a flight speed relative to the blade that is greater than the rotational speed of the blade, and the second UAV is used as a scanning UAV for the blade. The first UAV is used as a reference during the scanning process of the second UAV, and the laser ranging results for multiple positions on the blade are obtained through position data based on the first UAV as a reference and scanning data of the blade by the second UAV.

[0026] In one possible implementation, the fan blade detection method may further include:

[0027] The first UAV measures the rotation speed of the blade according to the flight speed and flight position of the blade followed by the first UAV, and controls the flight speed of the second UAV according to the rotation speed of the blade to keep the flight speed of the second UAV greater than the rotation speed of the blade;

[0028] The second UAV sends laser ranging data for scanning the blade to the first UAV;

[0029] The first UAV calculates through AI according to the laser ranging data and the rotation speed of the blade to obtain data of multiple position points on the surface of the blade.

[0030] In one possible implementation, the fan blade detection method may further include:

[0031] When the current battery level of the second UAV is less than a first preset battery level, the first UAV is used as a scanning UAV for the blade, and the second UAV is used as a reference for the first UAV during the scanning process. The first UAV is controlled to maintain a flight speed relative to the blade that is greater than the rotational speed of the blade, and the second UAV is controlled to fly at a speed that is the same as the rotational speed of the blade, so as to keep the second UAV and the blade stationary relative to each other.

[0032] The laser ranging results for multiple positions on the blade are obtained by using the position data based on the second UAV and the scanning data of the blade by the first UAV.

[0033] In one possible implementation, the fan blade detection method may further include:

[0034] When the difference between the current power levels of the first UAV and the second UAV is greater than a second preset power level, the first UAV is used as a scanning UAV for the blade, and the second UAV is used as a reference for the first UAV during the scanning process. The first UAV is controlled to maintain a flight speed relative to the blade that is greater than the rotational speed of the blade, and the second UAV is controlled to fly at a speed that is the same as the rotational speed of the blade, so as to keep the second UAV and the blade stationary relative to each other.

[0035] The laser ranging results for multiple positions on the blade are obtained by using the position data based on the second UAV and the scanning data of the blade by the first UAV.

[0036] According to some other exemplary embodiments of the present disclosure, a wind turbine blade detection device based on drone aerial photography is provided. The wind turbine blade detection device may include:

[0037] A first determination module is configured to determine a first initial three-dimensional position coordinate and posture of the wind turbine; wherein the first initial three-dimensional position coordinate is a coordinate in a GPS system, and the posture includes a direction of the wind turbine;

[0038] a second determining module, configured to determine, based on the first initial three-dimensional position coordinates, the posture, and a preset distance, a second initial three-dimensional coordinate of a first expected position of a front shooting position of the wind turbine; wherein the first expected position is located on a first normal vector corresponding to a first rotation center of a blade rotation plane of the wind turbine, and the distance between the first expected position and the first rotation center is the preset distance;

[0039] a first control module, configured to input second initial three-dimensional coordinates of the first expected position into a flight control system of the drone, and control the drone to fly to the second initial three-dimensional coordinates based on the GPS system; wherein the actual position of the drone is the actual three-dimensional coordinates, and the offset between the first expected position and the actual three-dimensional coordinates is a target offset vector;

[0040] a second control module, configured to control the drone to track at least one first key point of a blade of the wind turbine at the actual three-dimensional coordinate based on the orientation of the wind turbine, using a binocular camera in a first posture, and determine a three-dimensional motion trajectory of the first key point in a camera coordinate system;

[0041] a third determination module, configured to determine, in the camera coordinate system of the binocular camera, a second normal vector of a second rotation center corresponding to the three-dimensional motion trajectory of the first key point, convert the first normal vector into the camera coordinate system to obtain a third normal vector, and determine a first rotation matrix based on the second normal vector and the third normal vector;

[0042] an adjustment module, configured to adjust the orientation of the binocular camera based on the first rotation matrix so that the actual orientation of the binocular camera is parallel to the normal corresponding to the first rotation center of the actual wind blade rotation plane, obtain a second posture, and re-track at least one second key point of the wind blade using the binocular camera in the second posture to determine a three-dimensional motion trajectory of the second key point in the camera coordinate system and the re-determined second rotation matrix;

[0043] a judgment module, configured to judge whether the difference between the re-determined second rotation matrix and the first rotation matrix satisfies a preset requirement; if so, controlling the posture of the camera based on the orientation of the wind turbine and the second rotation matrix to capture a front image of the wind turbine; and if not, continuing iteration until the difference between two adjacent rotation matrices satisfies the preset requirement or the number of iterations reaches a preset number;

[0044] A third control module is used to determine the third initial three-dimensional coordinates of the second expected position for shooting the back of the wind turbine based on the first initial three-dimensional position coordinates, the posture and the preset distance, input the third initial three-dimensional coordinates of the second expected position into the flight control system of the UAV, control the UAV to fly to the optimized position corresponding to the third initial three-dimensional coordinates based on the GPS system, and control the camera to capture the back image of the wind turbine based on the orientation of the wind turbine and the second rotation matrix at the optimized position corresponding to the third initial three-dimensional coordinates.

[0045] According to some exemplary embodiments, a laser device may be provided on the drone; and the wind turbine blade detection device may further include:

[0046] a fourth control module, configured to control the UAV to fly around the blades of the wind turbine at all relative angles when the wind turbine is stationary, and to control the laser device to perform multiple laser ranging measurements on the blades by scanning during the flight around the blades, thereby obtaining multiple laser ranging results, and to establish a current model for the blades based on the multiple laser ranging results;

[0047] A comparison module, configured to compare the current model with a preset model to obtain a model comparison result; wherein the preset model is a model of the blades of the wind turbine in a normal state before operation;

[0048] A detection module is used to detect the degree of damage to the surface of the wind turbine blades according to the model comparison result.

[0049] According to some exemplary embodiments, the wind turbine blade detection device may further include: a fifth control module, used to control the drone to maintain a flight speed relative to the blade that is greater than the rotational speed of the blade when the wind turbine is in operation, so as to obtain laser ranging results for multiple positions on the blade.

[0050] Other exemplary embodiments of the present disclosure provide an electronic device, which may include a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and the processor implements the above-mentioned wind turbine blade detection method when executing the computer program.

[0051] In addition, some further exemplary embodiments of the present disclosure provide a computer-readable storage medium storing computer-executable instructions. When the computer-executable instructions are called and executed by a processor, the computer-executable instructions prompt the processor to execute the above-mentioned wind turbine blade detection method.

[0052] The embodiments of the present disclosure bring at least the following beneficial effects:

[0053] The embodiments of the present disclosure provide a method, device and electronic equipment for detecting wind turbine blades based on drone aerial photography, which can determine the first initial three-dimensional position coordinates and posture of the wind turbine; wherein the first initial three-dimensional position coordinates are coordinates in the GPS system, and the posture includes the orientation of the wind turbine; based on the first initial three-dimensional position coordinates, the posture and the preset distance, the second initial three-dimensional coordinates of the first expected position of the front shooting position of the wind turbine are determined; wherein the first expected position is located on the first normal vector corresponding to the first rotation center of the wind turbine blade rotation plane, and the distance between the first expected position and the first rotation center is the preset distance; the second initial three-dimensional coordinates of the first expected position are input into the flight control of the drone A control system, based on the GPS system, controls the drone to fly to the second initial three-dimensional coordinate; wherein the actual position of the drone is the actual three-dimensional coordinate, and the offset between the first expected position and the actual three-dimensional coordinate is the target offset vector; based on the orientation of the wind turbine, the drone is controlled at the actual three-dimensional coordinate, and at least one first key point of the wind blade of the wind turbine is tracked in a first posture by a binocular camera to determine the three-dimensional motion trajectory of the first key point in the camera coordinate system; a second normal vector of the second rotation center corresponding to the three-dimensional motion trajectory of the first key point is determined in the camera coordinate system of the binocular camera, the first normal vector is converted to the camera coordinate system to obtain a third normal vector, and based on the first normal vector The second normal vector and the third normal vector determine a first rotation matrix; based on the first rotation matrix, the orientation of the binocular camera is adjusted so that the actual orientation of the binocular camera is parallel to the normal corresponding to the first rotation center of the actual wind blade rotation plane, and a second posture is obtained, and the binocular camera is used to track at least one second key point of the wind blade in the second posture again to determine the three-dimensional motion trajectory of the second key point in the camera coordinate system and the re-determined second rotation matrix; it is judged whether the difference between the re-determined second rotation matrix and the first rotation matrix meets the preset requirements; if the preset requirements are met, the camera posture is controlled based on the orientation of the wind turbine and the second rotation matrix. the front image of the wind turbine is captured in an initial state; if the preset requirement is not met, the iteration is continued until the difference between two adjacent rotation matrices meets the preset requirement, or the number of iterations reaches the preset number; based on the first initial three-dimensional position coordinates, the posture and the preset distance, the third initial three-dimensional coordinates of the second expected position for the back shooting position of the wind turbine are determined, the third initial three-dimensional coordinates of the second expected position are input into the flight control system of the UAV, the UAV is controlled to fly to the optimized position corresponding to the third initial three-dimensional coordinates based on the GPS system, and the camera is controlled to capture the back image of the wind turbine based on the orientation of the wind turbine and the second rotation matrix at the optimized position corresponding to the third initial three-dimensional coordinates.In this solution, binocular vision technology can be used to adjust the posture of the shooting camera, and ultimately maintain a unified direction to capture images, greatly improving the image quality. In turn, when the model is used for recognition, the recognition accuracy can be improved, thereby improving the accuracy of blade surface damage detection.

[0054] In order to make the above-mentioned objectives, features and advantages of the present disclosure more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the specific embodiments of the present disclosure or the technical solutions in the related technologies, the following briefly introduces the drawings required for use in the specific embodiments or related technical descriptions. Obviously, the drawings described below are some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0056] FIG1 is a schematic diagram of a flow chart of a wind turbine blade detection method based on drone aerial photography provided by an embodiment of the present disclosure;

[0057] FIG2 is a schematic diagram of the relative positions between the expected position and the actual position provided by an embodiment of the present disclosure;

[0058] FIG3 is a schematic structural diagram of a wind turbine blade detection device based on drone aerial photography provided by an embodiment of the present disclosure;

[0059] FIG4 shows a schematic structural diagram of an electronic device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0060] To make the objectives, technical solutions, and advantages of the embodiments of the present disclosure more clear, the technical solutions of the present disclosure will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present disclosure, not all of them. Based on the embodiments of the present disclosure, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present disclosure.

[0061] The terms "including," "having," and any variations thereof, as used in the embodiments of the present disclosure, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or units is not limited to the listed steps or units, but may optionally include other steps or units not listed, or may optionally include other steps or units inherent to the process, method, product, or apparatus.

[0062] Currently, drone flight control and wind turbine positioning generally use the Global Positioning System (GPS) for positioning. Although using the same system can minimize inter-system errors, the limited accuracy of GPS itself causes a large difference between the actual and expected postures of drones when taking images, which significantly affects image quality and leads to low accuracy in blade surface damage detection.

[0063] Based on this, the embodiments of the present disclosure provide a wind turbine blade detection method, device, electronic device, and computer-readable storage medium based on drone aerial photography. This method can alleviate the technical problem of low accuracy in blade surface damage detection.

[0064] The embodiments of the present disclosure are further described below with reference to the accompanying drawings.

[0065] FIG1 is a flow chart of a wind turbine blade detection method based on drone aerial photography provided by an embodiment of the present disclosure. As shown in FIG1 , the method includes:

[0066] Step S110 , determining the first initial three-dimensional position coordinates and posture of the wind turbine.

[0067] The first initial three-dimensional position coordinates are coordinates in the GPS system, and the posture includes the direction of the wind turbine.

[0068] As a possible implementation, the first initial three-dimensional position coordinates and posture of the wind turbine are determined. The first initial three-dimensional position coordinates may be coordinates in a GPS system of a world coordinate system, and the posture may include the direction of the wind turbine.

[0069] Step S120 : determining a second initial three-dimensional coordinate of a first expected position of a front shooting position of the wind turbine based on the first initial three-dimensional position coordinate, the posture, and the preset distance.

[0070] The first expected position is located on a first normal vector corresponding to a first rotation center of a blade rotation plane of the fan, and a distance between the first expected position and the first rotation center is a preset distance.

[0071] For example, as shown in Figure 2, based on the first initial three-dimensional position coordinates, posture, and preset distance L of the wind turbine, the second initial three-dimensional coordinates of the expected position A of the front shooting position are determined. The expected position A is located on the normal line (first normal vector) corresponding to the rotation center of the wind blade rotation plane, and the distance between the expected position A and the rotation center B is the preset distance L.

[0072] In step S130 , the second initial three-dimensional coordinates of the first expected position are input into the flight control system of the UAV, and the UAV is controlled to fly to the second initial three-dimensional coordinates based on the GPS system.

[0073] The actual position of the UAV is the actual three-dimensional coordinate, and the offset between the first expected position and the actual three-dimensional coordinate is the target offset vector.

[0074] As an example, as shown in Figure 2, the second initial three-dimensional coordinates of the expected position A are input into the UAV flight control system, and the UAV is controlled to fly to the second initial three-dimensional coordinates based on the GPS system. Due to the system error of GPS, the actual position of the UAV is the actual three-dimensional coordinate A'; the offset between the expected position A and the actual three-dimensional coordinate A' is represented by the offset vector a1 (this value needs to be calculated).

[0075] Step S140 , based on the orientation of the wind turbine, the drone is controlled to track at least one first key point of the wind turbine blade at the actual three-dimensional coordinate using a binocular camera in a first posture to determine the three-dimensional motion trajectory of the first key point in the camera coordinate system.

[0076] For example, as shown in Figure 2, based on the orientation of the wind turbine, the drone is controlled at the actual three-dimensional coordinate A', and one or more first key points of the wind blade are tracked in a first posture through a binocular camera to determine the three-dimensional motion trajectory of the one or more first key points in the camera coordinate system (without scale information).

[0077] Step S150: Determine the second normal vector of the second rotation center corresponding to the three-dimensional motion trajectory of the first key point in the camera coordinate system of the binocular camera, convert the first normal vector to the camera coordinate system to obtain the third normal vector, and determine the first rotation matrix based on the second normal vector and the third normal vector.

[0078] In one possible implementation, in the camera coordinate system, the second normal vector of the rotation center B' corresponding to the three-dimensional motion trajectory of the one or more first key points is determined, then the first normal vector is converted to the camera coordinate system, the third normal vector is determined, and then the first rotation matrix is ​​determined based on the second normal vector and the third normal vector.

[0079] Step S160: Adjust the orientation of the binocular camera based on the first rotation matrix so that the actual orientation of the binocular camera is parallel to the normal corresponding to the first rotation center of the actual wind blade rotation plane, and obtain a second posture. Then, track at least one second key point of the wind blade again with the second posture through the binocular camera to determine the three-dimensional motion trajectory of the second key point in the camera coordinate system and the re-determined second rotation matrix.

[0080] In this step, the orientation of the binocular camera is first adjusted based on the rotation matrix (due to the existence of errors, the actual orientation of the camera is not parallel to the normal corresponding to the rotation center of the actual wind blade rotation plane during the first shooting. The purpose of the adjustment here is to make the actual orientation of the camera and the normal corresponding to the rotation center of the actual wind blade rotation plane more parallel) to obtain the second posture. Then, the binocular camera is used to track one or more second key points of the wind blade again in the second posture to determine the three-dimensional motion trajectory of the one or more second key points in the camera coordinate system (without scale information).

[0081] Step S170, determine whether the difference between the re-determined second rotation matrix and the first rotation matrix meets the preset requirements. If so, control the camera's posture based on the wind turbine's orientation and the second rotation matrix to capture a front image of the wind turbine. If not, continue iterating until the difference between two adjacent rotation matrices meets the preset requirements, or the number of iterations reaches the preset number.

[0082] In practical applications, the difference between the re-determined second rotation matrix and the first rotation matrix is ​​determined to determine whether it meets the preset requirements. If so, the posture of the high-speed camera is controlled based on the orientation of the wind turbine and the second rotation matrix to capture a front image of the wind turbine. If not, the iteration is continued until the difference between two adjacent rotation matrices meets the requirements, or the number of iterations reaches a preset number.

[0083] Step S180: Based on the first initial three-dimensional position coordinates, posture and preset distance, determine the third initial three-dimensional coordinates of the second expected position for shooting the back of the wind turbine, input the third initial three-dimensional coordinates of the second expected position into the flight control system of the UAV, control the UAV to fly to the optimized position corresponding to the third initial three-dimensional coordinates based on the GPS system, and control the camera to capture the back image of the wind turbine based on the orientation of the wind turbine and the second rotation matrix at the optimized position corresponding to the third initial three-dimensional coordinates.

[0084] As a possible implementation, the third initial three-dimensional coordinates of the expected position M for backside imaging are first determined based on the wind turbine's first initial three-dimensional position coordinates, its posture, and a preset distance L. Then, the initial three-dimensional coordinates of the preset position M are input into the drone's flight control system. The drone is controlled to fly to the optimized position M' based on the third initial three-dimensional coordinates using the GPS system. Then, at M', the high-speed camera is controlled to capture an image of the wind turbine's backside based on the wind turbine's orientation and the second rotation matrix. This allows for capturing images facing directly toward the wind turbine, although the wind turbine may not be located at the center of the image.

[0085] In the disclosed embodiment, binocular vision technology can be used to adjust the posture of the shooting camera, and ultimately the image can be captured in a uniform direction, which greatly improves the image quality. In turn, the recognition accuracy can be improved when the model is used for recognition, thereby improving the accuracy of blade surface damage detection.

[0086] The above steps are described in detail below.

[0087] In some embodiments, the drone is provided with a laser device; the method may further include the following steps:

[0088] Step a), when the wind turbine is stationary, controlling the drone to fly around the blades on the wind turbine at all relative angles, and controlling the laser device to perform multiple laser ranging on the blades by scanning during the flight around the blades, obtaining multiple laser ranging results, and establishing a current model for the blades based on the multiple laser ranging results;

[0089] Step b) comparing the current model with a preset model to obtain a model comparison result; wherein the preset model is a model of the blades of the wind turbine in a normal state before operation;

[0090] Step c) detecting the degree of surface damage of the wind turbine blades based on the model comparison results.

[0091] In the disclosed embodiment, through the laser ranging of the laser equipment on the drone and the flight of the drone around the blades on the wind turbine based on all relative angles, multiple laser ranging results for the blades can be obtained, and then the laser modeling process can be realized, so that the blade modeling method is not limited to the collected images, thereby improving the blade modeling efficiency.

[0092] Based on the above steps a), b) and c), the method may further include the following steps:

[0093] Step d), when the wind turbine is in operation, controlling the UAV to maintain a flight speed relative to the blades that is greater than the rotational speed of the blades, so as to obtain laser ranging results for multiple positions on the blades.

[0094] In the disclosed embodiment, for blades in motion, laser ranging can be performed on multiple parts of the blades by flying a drone at a speed greater than the blade rotational speed without affecting the normal operation of the wind turbine, making the blade detection process more efficient.

[0095] Based on the above step d), the number of drones is at least two; the method may further include the following steps:

[0096] Step e), when the wind turbine is in operation, controlling the first drone to fly at a speed equal to the rotational speed of the blades to keep the first drone and the blades relatively stationary;

[0097] Step f) controls the second UAV to maintain a flight speed relative to the blade that is greater than the rotational speed of the blade, and uses the second UAV as a scanning UAV for the blade, and uses the first UAV as a reference during the scanning process of the second UAV, and obtains laser ranging results for multiple positions on the blade through position data based on the first UAV as a reference and scanning data of the blade by the second UAV.

[0098] For example, when the wind turbine is running, the blades rotate. At this time, the first drone serving as a reference moves along with the rotation of the blades, so that the first drone and the blades are relatively stationary. The movement speed of the second drone serving as a scanning drone is different from the rotation speed of the blades, that is, the second drone moves relative to the blades, so that the second drone performs laser scanning on the blades and obtains laser ranging results at different positions on the blades.

[0099] In the embodiment of the present disclosure, by using the first drone as a reference, the scanning action of the second drone as a scanning drone can be based on the first drone, so that the laser scanning position of the blade is more accurate, and the data accuracy of the laser ranging result after laser scanning is improved.

[0100] Based on the above steps e) and f), the method may further include the following steps:

[0101] Step g), the first UAV measures the rotational speed of the blade according to the flight speed and flight position of the blade it is following, and controls the flight speed of the second UAV according to the rotational speed of the blade to keep the flight speed of the second UAV greater than the rotational speed of the blade;

[0102] Step h), the second UAV sends laser ranging data for scanning the blade to the first UAV;

[0103] In step i), the first UAV calculates the laser ranging data and the rotation speed of the blade through AI to obtain data of multiple position points on the surface of the blade.

[0104] For example, the first drone serving as a benchmark can also play the role of sharing the auxiliary computing pressure for the first drone. The first drone serving as a benchmark can also control the scanning action of the second drone, and then more accurately calculate the surface data of multiple position points of the blade through AI calculation based on the measured blade rotation speed.

[0105] In some embodiments, the method may further include the following steps:

[0106] Step j) When the current battery level of the second UAV is less than a first preset battery level, the first UAV is used as a scanning UAV for the blade, and the second UAV is used as a reference for the first UAV during the scanning process. The first UAV is controlled to maintain a flight speed relative to the blade that is greater than the rotational speed of the blade, and the second UAV is controlled to fly at a speed equal to the rotational speed of the blade, so that the second UAV and the blade remain stationary relative to each other.

[0107] Step k) is to obtain laser ranging results for multiple positions on the blade using the position data based on the second UAV and the scanning data of the blade by the first UAV.

[0108] It should be noted that since the second UAV serving as the scanning UAV has a relatively high movement speed and consumes relatively much power, when the power of the second UAV serving as the scanning UAV is low, the first UAV serving as the benchmark and the second UAV serving as the scanning UAV can switch with each other, that is, switch to the second UAV serving as the benchmark and the first UAV serving as the scanning UAV, so that the combination of the first UAV and the second UAV can continue the laser scanning ranging process for a longer period of time, making the laser scanning ranging process more efficient.

[0109] In some embodiments, the method may further include the following steps:

[0110] Step m): When the difference between the current power levels of the first UAV and the second UAV is greater than a second preset power level, the first UAV is used as a scanning UAV for the blade, and the second UAV is used as a reference for the first UAV during the scanning process. The first UAV is controlled to maintain a flight speed relative to the blade that is greater than the rotational speed of the blade, and the second UAV is controlled to fly at a speed equal to the rotational speed of the blade, so that the second UAV and the blade remain stationary relative to each other.

[0111] Step n): obtaining laser ranging results for multiple positions on the blade using the position data based on the second UAV and the scanning data of the blade by the first UAV.

[0112] As an optional implementation, when the difference in remaining power between the first drone serving as the benchmark and the second drone serving as the scanning drone is too large, they can switch to each other, that is, switch to the second drone serving as the benchmark and the first drone serving as the scanning drone. In this way, an intelligent power distribution process between multiple drones is realized, thereby improving the power utilization efficiency of drones during the laser scanning process.

[0113] FIG3 provides a schematic structural diagram of a wind turbine blade detection device based on drone aerial photography. As shown in FIG3 , the wind turbine blade detection device 300 based on drone aerial photography includes:

[0114] A first determination module 301 is configured to determine a first initial three-dimensional position coordinate and posture of the wind turbine; wherein the first initial three-dimensional position coordinate is a coordinate in a GPS system, and the posture includes the orientation of the wind turbine;

[0115] A second determining module 302 is configured to determine, based on the first initial three-dimensional position coordinates, the posture, and a preset distance, a second initial three-dimensional coordinate of a first expected position of a front shooting position of the wind turbine; wherein the first expected position is located on a first normal vector corresponding to a first rotation center of a blade rotation plane of the wind turbine, and the distance between the first expected position and the first rotation center is the preset distance;

[0116] A first control module 303 is configured to input the second initial three-dimensional coordinates of the first expected position into a flight control system of the UAV, and control the UAV to fly to the second initial three-dimensional coordinates based on the GPS system; wherein the actual position of the UAV is the actual three-dimensional coordinates, and the offset between the first expected position and the actual three-dimensional coordinates is a target offset vector;

[0117] A second control module 304 is configured to control the drone to track at least one first key point of a blade of the wind turbine at the actual three-dimensional coordinates based on the orientation of the wind turbine, using a binocular camera in a first posture, and determine a three-dimensional motion trajectory of the first key point in a camera coordinate system;

[0118] A third determination module 305 is configured to determine a second normal vector of a second rotation center corresponding to the three-dimensional motion trajectory of the first key point in the camera coordinate system of the binocular camera, convert the first normal vector into the camera coordinate system to obtain a third normal vector, and determine a first rotation matrix based on the second normal vector and the third normal vector;

[0119] An adjustment module 306 is configured to adjust the orientation of the binocular camera based on the first rotation matrix so that the actual orientation of the binocular camera is parallel to the normal corresponding to the first rotation center of the actual wind blade rotation plane, obtain a second posture, and re-track at least one second key point of the wind blade using the binocular camera in the second posture to determine a three-dimensional motion trajectory of the second key point in the camera coordinate system and the re-determined second rotation matrix;

[0120] A judgment module 307 is configured to determine whether the difference between the re-determined second rotation matrix and the first rotation matrix meets a preset requirement; if so, controlling the camera's posture based on the wind turbine's orientation and the second rotation matrix to capture a frontal image of the wind turbine; and if not, continuing the iteration until the difference between two adjacent rotation matrices meets the preset requirement or the number of iterations reaches a preset number.

[0121] The third control module 308 is used to determine the third initial three-dimensional coordinates of the second expected position for shooting the back of the wind turbine based on the first initial three-dimensional position coordinates, the posture and the preset distance, input the third initial three-dimensional coordinates of the second expected position into the flight control system of the UAV, control the UAV to fly to the optimized position corresponding to the third initial three-dimensional coordinates based on the GPS system, and control the camera to capture the back image of the wind turbine based on the orientation of the wind turbine and the second rotation matrix at the optimized position corresponding to the third initial three-dimensional coordinates.

[0122] In some embodiments, the drone is provided with a laser device; the device further comprises:

[0123] a fourth control module, configured to control the UAV to fly around the blades of the wind turbine at all relative angles when the wind turbine is stationary, and to control the laser device to perform multiple laser ranging measurements on the blades by scanning during the flight around the blades, thereby obtaining multiple laser ranging results, and to establish a current model for the blades based on the multiple laser ranging results;

[0124] A comparison module, configured to compare the current model with a preset model to obtain a model comparison result; wherein the preset model is a model of the blades of the wind turbine in a normal state before operation;

[0125] A detection module is used to detect the degree of damage to the surface of the wind turbine blades according to the model comparison result.

[0126] In some embodiments, the apparatus further comprises:

[0127] The fifth control module is used to control the UAV to maintain a flight speed relative to the blade that is greater than the rotation speed of the blade when the wind turbine is in operation, so as to obtain the laser ranging results for multiple positions on the blade.

[0128] In some embodiments, the number of the drones is at least two; the apparatus further comprises:

[0129] a sixth control module, configured to control the first UAV to fly at a speed equal to the rotational speed of the blades when the wind turbine is in operation, so as to keep the first UAV and the blades relatively stationary;

[0130] The seventh control module is used to control the second UAV to maintain a flight speed relative to the blade that is greater than the rotational speed of the blade, and to use the second UAV as a scanning UAV for the blade, and to use the first UAV as a reference during the scanning process of the second UAV, and to obtain the laser ranging results for multiple positions on the blade through the position data based on the first UAV as the reference and the scanning data of the second UAV for the blade.

[0131] In some embodiments, the apparatus further comprises:

[0132] a measurement module, configured to cause the first UAV to measure the rotational speed of the blade according to the flight speed and flight position of the blade followed by the first UAV, and to control the flight speed of the second UAV according to the rotational speed of the blade, so as to maintain the flight speed of the second UAV greater than the rotational speed of the blade;

[0133] A sending module, configured for the second UAV to send laser ranging data scanned for the blade to the first UAV;

[0134] A calculation module is used for the first UAV to calculate through AI based on the laser ranging data and the rotation speed of the blade to obtain data of multiple position points on the surface of the blade.

[0135] In some embodiments, the apparatus further comprises:

[0136] an eighth control module, configured to, when the current battery level of the second UAV is less than a first preset battery level, use the first UAV as a scanning UAV for the blade, use the second UAV as a reference for the first UAV during the scanning process, control the first UAV to maintain a flight speed relative to the blade that is greater than the rotational speed of the blade, and control the second UAV to fly at a speed equal to the rotational speed of the blade, so as to keep the second UAV and the blade stationary relative to each other;

[0137] The first acquisition module is configured to acquire the laser ranging results for a plurality of positions on the blade using the position data based on the second UAV and the scanning data of the blade by the first UAV.

[0138] In some embodiments, the apparatus further comprises:

[0139] a ninth control module, configured to, when the difference between the current power levels of the first and second drones is greater than a second preset power level, use the first drone as a scanning drone for the blade, use the second drone as a reference for the first drone's scanning process, control the first drone to maintain a flight speed relative to the blade that is greater than a rotational speed of the blade, and control the second drone to fly at a speed equal to the rotational speed of the blade, so as to keep the second drone stationary relative to the blade;

[0140] The second acquisition module is used to obtain the laser ranging results for multiple positions on the blade by using the position data based on the second UAV and the scanning data of the blade by the first UAV.

[0141] The wind turbine blade detection device based on drone aerial photography provided in the embodiment of the present disclosure has the same technical features as the wind turbine blade detection method based on drone aerial photography provided in the above embodiment, so it can also solve the same technical problems and achieve the same technical effects.

[0142] An electronic device provided by an embodiment of the present disclosure, as shown in FIG4 , includes a processor 402 and a memory 401 , wherein the memory stores a computer program that can be run on the processor, and when the processor executes the computer program, the steps of the method provided in the above embodiment are implemented.

[0143] 4 , the electronic device further includes a bus 403 and a communication interface 404 . A processor 402 , the communication interface 404 , and the memory 401 are connected via the bus 403 . The processor 402 is configured to execute executable modules stored in the memory 401 , such as computer programs.

[0144] Memory 401 may include high-speed random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage. Communication between the system network element and at least one other network element is achieved through at least one communication interface 404 (which may be wired or wireless), and may use the Internet, a wide area network, a local area network, a metropolitan area network, etc.

[0145] Bus 403 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, FIG4 shows only one bidirectional arrow, but this does not mean that there is only one bus or one type of bus.

[0146] Among them, the memory 401 is used to store programs, and the processor 402 executes the program after receiving the execution instruction. The method executed by the device defined by the process disclosed in any embodiment of the present disclosure can be applied to the processor 402 or implemented by the processor 402.

[0147] The processor 402 may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be completed by the hardware integrated logic circuit in the processor 402 or by instructions in the form of software. The above-mentioned processor 402 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gates or transistor logic devices, or discrete hardware components. The various methods, steps, and logic block diagrams disclosed in the embodiments of the present disclosure can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. The steps of the method disclosed in conjunction with the embodiments of the present disclosure can be directly embodied as being executed by a hardware decoding processor, or can be executed by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium well-known in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in memory 401, and processor 402 reads the information in memory 401 and, in conjunction with its hardware, completes the steps of the above method.

[0148] Corresponding to the above-mentioned wind turbine blade detection method based on drone aerial photography, an embodiment of the present disclosure also provides a computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are called and executed by the processor, the computer-executable instructions prompt the processor to execute the steps of the above-mentioned wind turbine blade detection method based on drone aerial photography.

[0149] The wind turbine blade detection device based on drone aerial photography provided by the embodiment of the present disclosure can be specific hardware on the equipment or software or firmware installed on the equipment. The implementation principle and technical effects of the device provided by the embodiment of the present disclosure are the same as those of the aforementioned method embodiment. For the sake of brief description, for matters not mentioned in the device embodiment, reference can be made to the corresponding content in the aforementioned method embodiment. Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can all refer to the corresponding processes in the aforementioned method embodiment, and will not be repeated here.

[0150] In the embodiments provided in the present disclosure, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interface, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0151] For another example, the flowcharts and block diagrams in the accompanying drawings show the possible architectures, functions and operations of the devices, methods and computer program products according to multiple embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of code, and the module, program segment or a part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0152] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0153] In addition, each functional unit in the embodiments provided in the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0154] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present disclosure, or the part that contributes to the relevant technology or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the wind turbine blade detection method based on drone aerial photography described in each embodiment of the present disclosure. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, and other media that can store program codes.

[0155] It should be noted that similar numbers and letters represent similar items in the following figures. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. In addition, the terms "first", "second", "third", etc. are only used to distinguish the description and are not to be understood as indicating or implying relative importance.

[0156] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present disclosure, which are used to illustrate the technical solutions of the present disclosure, rather than to limit them. The protection scope of the present disclosure is not limited thereto. Although the present disclosure has been described in detail with reference to the above-mentioned embodiments, ordinary technicians in this field should understand that any technician familiar with this technical field can still modify the technical solutions described in the above-mentioned embodiments within the technical scope disclosed in the present disclosure, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present disclosure. They should all be included in the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be based on the protection scope of the claims. Industrial Applicability

[0157] The present disclosure provides a method, device, electronic device, and computer-readable storage medium for detecting wind turbine blades based on drone aerial photography, which relates to the field of power equipment technology and alleviates the technical problem of low accuracy in detecting damage on the blade surface. The method includes: determining a first initial three-dimensional position coordinate and posture of the wind turbine; determining a second initial three-dimensional coordinate of a first expected position for a frontal shooting position of the wind turbine based on the first initial three-dimensional position coordinate, posture, and a preset distance; inputting the second initial three-dimensional coordinate of the first expected position into a flight control system of the drone, and controlling the drone to fly to the second initial three-dimensional coordinate based on a GPS system; controlling the drone to track at least one first key point of the wind turbine blade at the actual three-dimensional coordinate based on the orientation of the wind turbine using a binocular camera in a first posture, and determining the three-dimensional motion trajectory of the first key point in the camera coordinate system.

[0158] Furthermore, it is understood that the wind turbine blade inspection method, apparatus, and electronic device disclosed herein based on drone aerial photography are reproducible and applicable to a variety of industrial applications. For example, the wind turbine blade inspection method disclosed herein can be applied to the field of power equipment.

Claims

1. A wind turbine blade detection method based on drone aerial photography, wherein: The fan blade detection method comprises: Determine a first initial three-dimensional position coordinate and posture of the wind turbine; wherein the first initial three-dimensional position coordinate is a coordinate in a GPS system, and the posture includes a direction of the wind turbine; Based on the first initial three-dimensional position coordinates, the posture and the preset distance, determine the second initial three-dimensional coordinates of the first expected position of the front shooting position of the wind turbine; wherein the first expected position is located on the first normal vector corresponding to the first rotation center of the wind blade rotation plane of the wind turbine, and the distance between the first expected position and the first rotation center is the preset distance; Inputting the second initial three-dimensional coordinates of the first expected position into a flight control system of the UAV, and controlling the UAV to fly to the second initial three-dimensional coordinates based on the GPS system; wherein the actual position of the UAV is the actual three-dimensional coordinates, and the offset between the first expected position and the actual three-dimensional coordinates is the target offset vector; Based on the orientation of the wind turbine, the drone is controlled at the actual three-dimensional coordinate, and at least one first key point of a blade of the wind turbine is tracked in a first posture by a binocular camera to determine a three-dimensional motion trajectory of the first key point in a camera coordinate system; Determine a second normal vector of a second rotation center corresponding to the three-dimensional motion trajectory of the first key point in the camera coordinate system of the binocular camera, convert the first normal vector to the camera coordinate system to obtain a third normal vector, and determine a first rotation matrix based on the second normal vector and the third normal vector; The orientation of the binocular camera is adjusted based on the first rotation matrix so that the actual orientation of the binocular camera is parallel to the normal corresponding to the first rotation center of the actual wind blade rotation plane, and a second posture is obtained. The binocular camera is used to track at least one second key point of the wind blade in the second posture again, and the three-dimensional motion trajectory of the second key point in the camera coordinate system and the re-determined second rotation matrix are determined; Determine whether the difference between the re-determined second rotation matrix and the first rotation matrix meets the preset requirements; if the preset requirements are met, control the posture of the camera based on the orientation of the wind turbine and the second rotation matrix to capture the front image of the wind turbine; if the preset requirements are not met, continue to iterate until the difference between two adjacent rotation matrices meets the preset requirements, or the number of iterations reaches the preset number; Based on the first initial three-dimensional position coordinates, the posture and the preset distance, determine the third initial three-dimensional coordinates of the second expected position for shooting the back side of the wind turbine, input the third initial three-dimensional coordinates of the second expected position into the flight control system of the UAV, control the UAV to fly to the optimized position corresponding to the third initial three-dimensional coordinates based on the GPS system, and control the camera to capture the back image of the wind turbine based on the orientation of the wind turbine and the second rotation matrix at the optimized position corresponding to the third initial three-dimensional coordinates.

2. The wind turbine blade detection method according to claim 1, wherein: The UAV is provided with a laser device; the fan blade detection method further comprises: When the wind turbine is in a stationary state, the drone is controlled to fly around the blades on the wind turbine based on all relative angles, and in the process of flying around the blades, the laser device is controlled to always perform multiple laser distance measurements on the blades by scanning, so as to obtain multiple laser distance measurement results, and a current model for the blades is established according to the multiple laser distance measurement results; Comparing the current model with a preset model to obtain a model comparison result; wherein the preset model is a model of the blades of the wind turbine in a normal state before operation; The degree of surface damage of the fan blades is detected according to the model comparison result.

3. The wind turbine blade detection method according to claim 2, wherein: The fan blade detection method also includes: When the wind turbine is in operation, the drone is controlled to maintain a flight speed relative to the blade that is greater than a rotation speed of the blade, so as to obtain the laser ranging results for multiple positions on the blade.

4. The wind turbine blade detection method according to claim 3, wherein: The number of the drones is at least two; The fan blade detection method also includes: When the wind turbine is in operation, controlling the first drone to fly at a speed equal to the rotation speed of the blades to keep the first drone and the blades relatively stationary; The second UAV is controlled to maintain a flight speed relative to the blade that is greater than the rotational speed of the blade, and the second UAV is used as a scanning UAV for the blade. The first UAV is used as a reference during the scanning process of the second UAV, and the laser ranging results for multiple positions on the blade are obtained through position data based on the first UAV as a reference and scanning data of the blade by the second UAV.

5. The wind turbine blade detection method according to claim 4, wherein: The fan blade detection method also includes: The first UAV measures the rotation speed of the blade according to the flight speed and flight position of the blade followed by the first UAV, and controls the flight speed of the second UAV according to the rotation speed of the blade to keep the flight speed of the second UAV greater than the rotation speed of the blade; The second UAV sends laser ranging data for scanning the blade to the first UAV; The first drone calculates through AI according to the laser ranging data and the rotation speed of the blade to obtain data of multiple position points on the surface of the blade.

6. The wind turbine blade detection method according to claim 4 or 5, wherein: The fan blade detection method also includes: When the current power of the second drone is less than the first preset power, the first drone is used as a scanning drone for the blade, the second drone is used as a reference during the scanning process of the first drone, and the first drone is controlled to maintain a flight speed greater than the rotation speed of the blade to fly, and the second drone is controlled to maintain a flight speed greater than the rotation speed of the blade. The second drone flies at a speed equal to the rotation speed of the blades to keep the second drone and the blades relatively stationary; The laser ranging results for multiple positions on the blade are obtained by using the position data based on the second UAV and the scanning data of the blade by the first UAV.

7. The wind turbine blade detection method according to claim 4 or 5, wherein: The fan blade detection method also includes: When the current power difference between the first UAV and the second UAV is greater than a second preset power, the first UAV is used as a scanning UAV for the blade, the second UAV is used as a reference in the scanning process of the first UAV, and the first UAV is controlled to maintain a flight speed for the blade greater than the rotation speed of the blade, and the second UAV is controlled to fly at a speed equal to the rotation speed of the blade, so as to keep the second UAV and the blade relatively stationary; The laser ranging results for multiple positions on the blade are obtained by using the position data based on the second UAV and the scanning data of the blade by the first UAV.

8. A wind turbine blade detection device based on drone aerial photography, wherein: The fan blade detection device comprises: A first determination module is configured to determine a first initial three-dimensional position coordinate and a posture of the wind turbine; wherein the first initial three-dimensional position coordinate is a coordinate in a GPS system, and the posture includes a direction of the wind turbine; A second determination module is configured to determine a second initial three-dimensional coordinate of a first expected position of a front shooting position of the wind turbine based on the first initial three-dimensional position coordinate, the posture, and a preset distance; wherein the first expected position is located on a first normal vector corresponding to a first rotation center of a blade rotation plane of the wind turbine, and a distance between the first expected position and the first rotation center is the preset distance; a first control module, configured to input the second initial three-dimensional coordinates of the first expected position into a flight control system of the drone, and control the drone to fly to the second initial three-dimensional coordinates based on the GPS system; wherein the actual position of the drone is the actual three-dimensional coordinates, and the offset between the first expected position and the actual three-dimensional coordinates is a target offset vector; A second control module is configured to control the drone at the actual three-dimensional coordinate based on the orientation of the wind turbine, track at least one first key point of the wind blade of the wind turbine in a first posture through a binocular camera, and determine a three-dimensional motion trajectory of the first key point in a camera coordinate system; A third determination module is configured to determine a second normal vector of a second rotation center corresponding to the three-dimensional motion trajectory of the first key point in the camera coordinate system of the binocular camera, convert the first normal vector into the camera coordinate system to obtain a third normal vector, and determine a first rotation matrix based on the second normal vector and the third normal vector; an adjustment module, configured to adjust the orientation of the binocular camera based on the first rotation matrix so that the actual orientation of the binocular camera is parallel to the normal corresponding to the first rotation center of the actual wind blade rotation plane, obtain a second posture, and re-adjust at least one of the wind blades through the binocular camera in the second posture; A second key point is tracked to determine a three-dimensional motion trajectory of the second key point in the camera coordinate system and a re-determined second rotation matrix; A judgment module is configured to judge whether the difference between the re-determined second rotation matrix and the first rotation matrix meets a preset requirement; if the preset requirement is met, the front image of the wind turbine is captured by controlling the posture of the camera based on the orientation of the wind turbine and the second rotation matrix; if the preset requirement is not met, the iteration is continued until the difference between two adjacent rotation matrices meets the preset requirement or the number of iterations reaches a preset number; The third control module is configured to determine the third initial three-dimensional coordinates of the second expected position for shooting the back side of the wind turbine based on the first initial three-dimensional position coordinates, the posture and the preset distance, input the third initial three-dimensional coordinates of the second expected position into the flight control system of the UAV, control the UAV to fly to the optimized position corresponding to the third initial three-dimensional coordinates based on the GPS system, and control the camera to capture the back image of the wind turbine based on the orientation of the wind turbine and the second rotation matrix at the optimized position corresponding to the third initial three-dimensional coordinates.

9. The wind turbine blade detection device based on drone aerial photography according to claim 8, wherein: The UAV is provided with a laser device; the fan blade detection device also includes: a fourth control module, configured to control the UAV to fly around the blades on the wind turbine at all relative angles when the wind turbine is in a stationary state, and to control the laser device to always perform multiple laser distance measurements on the blades by scanning in the process of flying around the blades, to obtain multiple laser distance measurement results, and to establish a current model for the blades according to the multiple laser distance measurement results; A comparison module, configured to compare the current model with a preset model to obtain a model comparison result; wherein the preset model is a model of the blades of the wind turbine in a normal state before operation; The detection module is configured to detect the degree of damage on the surface of the wind turbine blades according to the model comparison result.

10. The wind turbine blade detection device based on drone aerial photography according to claim 9, wherein: The fan blade detection device also includes: The fifth control module is configured to control the UAV to maintain a flight speed relative to the blade that is greater than the rotation speed of the blade when the wind turbine is in operation, so as to obtain the laser ranging results for multiple positions on the blade.

11. An electronic device comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, wherein: When the processor executes the computer program, the steps of the wind turbine blade detection method described in any one of claims 1 to 7 are implemented.

12. A computer-readable storage medium, wherein: The computer-readable storage medium stores computer-executable instructions. When the computer-executable instructions are called and executed by a processor, the computer-executable instructions prompt the processor to execute the wind turbine blade detection method according to any one of claims 1 to 7.

Images